Dissertation/Thesis Abstract

Permian Basin Reservoir Quantitative Interpretation Applying the Multi-Scale Boxcar Transform Spectral Decomposition
by Locci-Lopez, Daniel Eduardo, M.S., University of Louisiana at Lafayette, 2018, 74; 10816133
Abstract (Summary)

The Short Time Fourier transform and the S-transform are among the most used methods of spectral decomposition to localize spectra in time and frequency. The S-transform utilizes a frequency-dependent Gaussian analysis window that is normalized for energy conservation purposes. The STFT, on the other hand, has a selected fixed time window that does not depend on frequency. In previous literature, it has been demonstrated that the S-transform distorts the Fourier spectra, shifting frequency peaks, and could result in misleading frequency attributes. Therefore, one way of making the S-transform more appropriate for quantitative seismic signal analysis is to ignore the conservation of energy over time requirement. This suggests a hybrid approach between the Short Time Fourier transform and the S-transform for seismic interpretation purposes.

In this work, we introduce the Multi-Scale Boxcar transform that has temporal resolution comparable to the S-transform while giving correct Fourier peak frequencies. The Multi-Scale Boxcar transform includes a special analysis window that focusses the analysis on the highest amplitude portion of the Gaussian window, giving a more accurate time-frequency representation of the spectra in comparison with the S-transform.

Post-stack seismic data with a strong well logs control was used to demonstrate the differences of the Multi-Scale Boxcar transform and the S-transform. The analysis area in this work is the Pennsylvanian and Lower Permian Horseshoe Atoll Carbonate play in the Midland Basin, a sub-basin in the larger Permian Basin. The Multi-Scale Boxcar transform spectral decomposition method improved the seismic interpretation of the study area, showing better temporal resolution for resolving the layered reservoirs’ heterogeneity. The time and depth scale values on the figures are shifted according to the sponsor request, but the relative scale is correct.

Indexing (document details)
Advisor: Zhang, Rui, Castagna, John
Commitee: Gottardi, Raphael, Morra, Gabriele, Willis, James
School: University of Louisiana at Lafayette
Department: Geology
School Location: United States -- Louisiana
Source: MAI 58/05M(E), Masters Abstracts International
Subjects: Computational physics, Geophysics, Petroleum Geology
Keywords: Horseshoe Atoll Carbonate play, Multi-Scale Boxcar transform, Permian Basin, Quantitative interpretation, S-transform, Spectral decomposition
Publication Number: 10816133
ISBN: 9781392041475
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy